Over the past few years, an impressive array of active matrix liquid crystal display (AMLCD) products have been developed that demonstrate very high-quality display performance. There is intense interest in and demand for such displays for computer, avionic, automobile, and consumer products. The growth of the AMLCD market will be controlled by manufacturing capacity and cost. In the second part of the 1980s, many Japanese companies (display manufacturers and materials and equipment suppliers) made plans for large financial commitments to this technology. Now, in the early 1990s, very large manufacturing facilities are being built to meet the enormous demand for AMLCDs. The initial largest product volume is in color video graphic adapter (VGA) displays for the laptop computer market. With known committed investments of over $2 billion, there is considerable pressure to quickly achieve high manufacturing yield and low cost and obtain an acceptable return on investment. Manufacturing issues have become the prime focus of research and development. There is intense competition for market share, because many major Japanese corporations view this area as a strategic long-term investment.
AMLCD manufacturing technology is unique, but it has some similarities to integrated circuit (IC) and conventional LCD processes. Although all the major research discoveries in LCD materials and applications were made in Europe and the United States, Japan has dominated the world production of twisted nematic and supertwist displays. Many of the techniques used in cell assembly in these technologies are similar to those used in AMLCD. The dominant active matrix technology is thin-film transistors (TFTs) of either a-Si or p-Si. The lower process temperature (<400 degrees centigrade) of a-Si has led to its initial dominance in large-area AMLCD technology. There is in Japan substantial experience and infrastructure in large-area a-Si manufacturing for solar cells, which was initiated by the Ministry of International Trade and Industry (MITI) during the oil crisis of the 1970s. Some of this experience, in particular large-area a-Si deposition, has been useful, because there are now in Japan established suppliers of large-area plasma-enhanced CVD (PECVD) manufacturing equipment. The present manufacturing of p-Si LCD displays is confined to small-size quality substrates (<6 inches) using high- temperature "IC-like" processes. Most of these products are for small high-density camcorder and projection applications. Low-temperature large-area p-Si device and process development are still in the research stage in Japan. While there is considerable interest in developing low- temperature p-Si AMLCDs because it would be possible to incorporate p-Si built- in drivers in the display, most Japanese companies have had to devote their resources to making the manufacturing of a-Si AMLCDs successful.
There are a considerable number of publications that describe device architectures and display performance, but manufacturing processes and yield issues naturally remain as company secrets in this intensely competitive market. This chapter describes some of the a-Si AMLCD factory logistics and discusses some aspects of throughput yield and defects that have been deduced from many visits and discussions with a considerable number of manufacturers. Some major manufacturing equipment is described, along with a number of other infrastructure issues. Given some of the sensitive issues surrounding manufacturing of AMLCDs, precise information is not directly available; thus, some of our information is based on incomplete data.